"That" being the term "media bias".
I have a lot of comments on an interesting post by David Henderson on "media bias" and a new book on the subject by Tim Groseclose. It was similar in a lot of ways to my interest in his earlier article with Gochenour on presidential greatness: great subject discussed by a guy I consider both sharp and fair, but a deeply problematic argument.
The problem with Henderson and Groseclose's discussion of "media bias" (aside from a few bad variable choices - see my comments on that) is that the case boils down to something like "the media is biased because it is more liberal than the average American and a lot more liberal than me". If this is how we're talking about "bias" now, we're in trouble. I don't know where this view comes from that it's the media that's biased if it doesn't reflect the population. Maybe the population is biased. We are too willing to use the term "unbiased" to mean that both sides of the debate get equal air time. It's hogwash. Of course truth can be contested - most definitely. And as a guy that leans toward pragmatist explanations of these things I'm more than willing to say that. But the fact that truth is contested does not mean that truth is the average of what the population thinks about something. Truth is contested because of our observational, analytical, and computational frailties - not because truth is a mirror of all of our views.
What's most amazing to me is that Henderson and Groseclose use the distribution of the political views of Congress as the yardstick to judge the bias of the media. That seems almost exactly backward to me. Why don't we instead say "the media's job is to report the facts - they are obviously not perfect at it, but since they are the fact-reporters perhaps we should judge the bias of Congress by their conformity to the media". If Groseclose and Henderson took that approach, instead of concluding that the media has a liberal bias because they don't reproduce the political pathologies of the Congress, they would have concluded that Congress has a conservative/libertarian bias.
I didn't plan on sharing my thoughts on Henderson's review here, but this post by Krugman inspired me to. It's a great example of a case where there's a real divergence between media positions and political positions, but I think it's fair to say that it's the politicians that are biased - not the media - and if you judge the media by their conformity to the political discourse you get exactly the opposite answer that you should get (if you're concerned about bias).
So what's the source of this argument? Confirmation bias? Would Groseclose and Henderson be arguing these positions if it cut the other way? Would I?